Pre-installation visual inspection

Pre-installation visual inspection means checking modules by eye before they are mounted. It is listed as a preventive measure against backsheet cracking. It is recommended together with continuous ground fault detection and a check that the IEC 61215 certification and bill of materials are valid. Backsheet cracking can start during production or installation, so checking modules before mounting can catch defects before they reach the field. Visual inspection is described as the quickest and most effective way to find module defects. It works well for burn marks, delamination, encapsulant yellowing, backsheet blistering and junction box failure. Look for bent or misaligned frames, surfaces or junction boxes, cracked cells and delamination, because these defects can lead to later power loss or safety failures. The IEC-based method splits the module into its parts and inspects each one under good lighting, counting only defects visible to the bare eye. Record the position and size of every defect and take photos without reflections. Its limit is that many performance changes cannot be seen and need more advanced tools to detect.

<!-- review notes

• sentence 1: support check: partial - Passages list visual inspection before installation as a measure for backsheet cracking, but do not define it as checking modules by eye before mounting (only implied by name).
• sentence 3: support check: partial - Passages list production and installation as phases of backsheet cracking, but do not state the inference that pre-mounting checks catch defects before the field.
• sentence 4: support check: partial - Passages say visual inspection is the most effective and quickest method and effective for burn marks, delamination, yellowing, blistering, and junction box failure, but 'works well' is stated generally and the quickest claim is about finding failures; mostly supported, though the u4 text also says it is poorly applicable to weathered modules.
• sentence 6: support check: partial - Parts division, bare-eye defects, position/size recording, and reflection-free photos are stated, but 'good lighting' is specified as >1000 lux and the method is described as lab practice; 'counting only' and 'every defect' are close enough but the core is supported. -->